Real-Time Doctor-Patient Consultations Website

This project is a Real-Time Doctor-Patient consultations website designed to facilitate secure, real-time text-based consultations between doctors and patients. With a focus on simplicity and ease of use, the website allows patients and doctors to engage in private, secure conversations without the need for in-person visits, leveraging modern web technologies.

Live Website You can try the website live at: Live Website Link

Table of Contents

• Features
• What's next?
• Screenshots
• Roadmap
• Technologies Used
• Run the project
• Contributing
• License

Features

• Real-time Secure Video Calls: Private, text-based messaging + video calls for doctor-patient communication.
• AI Doctor assistant for patients that is conversation aware and supports PDF files.
• Appointment Scheduling: Patients can book consultations with doctors, ensuring a smooth experience.
• User Profiles: Both doctors and patients have personalized profiles for storing essential information.
• Real-time Notifications: Receive notifications when your consultation is about to begin, or when a doctor is available.
• Responsive Design: Accessible on all devices, ensuring users can connect from anywhere.

What's next?

• Add medical knowledge layer (RAG) Use medical sources:

  • WHO
  • Mayo Clinic
  • NHS

• Safety layer (VERY important for medical AI) filter out:

  • emergency symptoms (chest pain, stroke signs, etc.)
  • self-harm / dangerous behavior
  • medication requests

Then override behavior:

Example rule:

If “chest pain + shortness of breath” → always recommend emergency care first

• tool use : extend it with:

  • symptom checker flow (decision tree)
  • drug database lookup
  • lab value explanation tool

• . AI evaluation system (this is VERY AI-engineering heavy)

  Create a test suite:

  Example test cases:

  chest pain headache fever anxiety symptoms

  Evaluate:

  did model suggest doctor visit? did it hallucinate diagnosis? response quality score

  👉 This teaches:

  LLM evaluation prompt testing regression testing for AI

• ⚠️ Safety + Triage Classifier Layer (Pre-LLM logic)

• 🧾 Structured Clinical Output Generator (Function-calling style)

Screeshots

Landing Page

Doctor Dashboard

Patient Home Page

Book a Consultation

Real-time Consultation

AI Consultation

Roadmap

• Implement video consultation functionality (coming soon).
• Integrate payment processing for paid consultations.
• Add multi-language support for international users.

Technologies Used

• Frontend: NextJS, Shadcn
• Backend: Node.js, Express.js, MongoDB Atlas
• AI: Gemini API, Vercel AI-SDK
• Authentication: Firebase
• Real-time chat: Socket.io
• Real-time notifications: Node Cron
• Deployment: Render.com, Netlify

Run the project

Clone the repository

git clone https://github.com/aminezouari52/telemedicine-website.git

Create the .env files

• navigate to the apps/backend and apps/frontend folders
• use .env.example as an example

Navigate to the directory

cd telemedicine-website

Use the package manager pnpm to install and run the project.

pnpm install
pnpm dev

Contributing

This project is open-source, and I’m excited to collaborate with developers around the world.

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

MIT